Key points

Introduction

Recent genomic findings in head and neck squamous cell carcinoma (HNSCC) reveal a wide spectrum of genomic alterations. A heterogeneous disease by nature, HNSCC encompasses a disparate collection of anatomic sites with complex tumor biology. One of the most distinguishing features of HNSCC is the human papillomavirus (HPV) status of the tumor, HPV-positive HNSCC having more favorable outcomes compared with HPV-negative HNSCC. This genomic heterogeneity of HNSCC tumors creates an obstacle for the identification of an effective targeting agent likely to benefit most patients with HNSCC. The most successful implementation of the genomic alterations in recent years has been based on functionally activating gene mutations (eg, c-KIT activating mutations in gastrointestinal stromal tumor ) and copy number gain in oncogenes (eg, ERBB2/HER2 amplification in breast cancer ).

However, based on current genome-wide sequencing and copy number data, there are only a few oncogenes in HNSCC that can be immediately exploited with novel targeted agents. This article focuses on novel approaches targeting potentially critical pathways that are frequently altered in HNSCC, including DNA damage response and cell cycle regulation, PI3K/mTOR, NOTCH, transmembrane growth factor receptors (c-MET, FGFR, and Axl), and angiogenesis. The established therapeutic target inhibitors including epidermal growth factor receptor (EGFR) and immune check point proteins (eg, PD1 and PD-L1) are discussed elsewhere in this issue.

Introduction

Recent genomic findings in head and neck squamous cell carcinoma (HNSCC) reveal a wide spectrum of genomic alterations. A heterogeneous disease by nature, HNSCC encompasses a disparate collection of anatomic sites with complex tumor biology. One of the most distinguishing features of HNSCC is the human papillomavirus (HPV) status of the tumor, HPV-positive HNSCC having more favorable outcomes compared with HPV-negative HNSCC. This genomic heterogeneity of HNSCC tumors creates an obstacle for the identification of an effective targeting agent likely to benefit most patients with HNSCC. The most successful implementation of the genomic alterations in recent years has been based on functionally activating gene mutations (eg, c-KIT activating mutations in gastrointestinal stromal tumor ) and copy number gain in oncogenes (eg, ERBB2/HER2 amplification in breast cancer ).

However, based on current genome-wide sequencing and copy number data, there are only a few oncogenes in HNSCC that can be immediately exploited with novel targeted agents. This article focuses on novel approaches targeting potentially critical pathways that are frequently altered in HNSCC, including DNA damage response and cell cycle regulation, PI3K/mTOR, NOTCH, transmembrane growth factor receptors (c-MET, FGFR, and Axl), and angiogenesis. The established therapeutic target inhibitors including epidermal growth factor receptor (EGFR) and immune check point proteins (eg, PD1 and PD-L1) are discussed elsewhere in this issue.

DNA damage response and cell cycle regulation targeted agents

Targeting Dysfunctional p53

The gene TP53 (tumor protein p53, p53 , or Trp53 ), was in 1979 the first tumor-suppressor gene to be identified, and the protein product of this gene, p53, is one of the most important molecules in biology, integrating numerous signals that control cell cycling and apoptosis. The p53 network is normally inactive and responds to stimuli, such as cellular damage or stress, to perform its tumor-suppressing function. There are many ways in which p53 protein malfunctions in human cancers, and at least six different mechanisms are cited: (1) amino-acid-changing mutation in the DNA binding domain; (2) deletion of the carboxy-terminal domain; (3) multiplication of the MDM2 gene in the genome; (4) viral infection; (5) deletion of the p14ARF gene; and (6) mislocalization of p53 to the cytoplasm, outside the nucleus.

In HNSCC, the cause of p53 dysfunction differs between HPV-positive and HPV-negative tumors. The HPV-positive HNSCC lacks functional p53 because ubiquitination of p53 by an ubiquitin ligase, E6AP, and a viral oncoprotein, E6, leads to rapid degradation of p53, whereas TP53 mutation is very rare. However, TP53 is the most frequently mutated gene in HPV-negative HNSCC, with an incidence of 47% to 87%. These mutations can occur throughout the entire gene and typically involve missense mutations, which alter protein conformation or affect how p53 binds its DNA targets. It is also known that any TP53 mutations in tumor DNA are associated with reduced survival after surgical treatment of HNSCC.

Given the biologic importance of p53, there has been a significant effort to identify effective therapeutic approaches for HNSCC with p53 dysfunction. However, direct targeting of a tumor suppressor, such as p53, is currently not feasible because restoration of a lost protein function in the appropriate cellular regulatory context is difficult compared with the inhibition of overly active proteins, such as deregulated oncoproteins. One of the ways to circumvent this problem is to find a synthetic lethal partner for dysfunctional p53. The synthetic lethality therapeutic approaches consist of a combination of two or more separate genes/proteins’ functional loss leading to cell death, whereas functional loss of only one of the genes/proteins does not reduce cell viability.

Moser and colleagues applied a functional kinomic approach using a high-throughput RNA interference platform to identify new targets exploiting dependence on G2-M cell cycle regulators of TP53 -mutant tumors for their viability ( Fig. 1 A and B). WEE1 (a G2-M cell cycle regulator), CAM2KB, and NEK4 were identified as the most promising candidate target kinases. MK-1775 (AZD1775), the WEE1 kinase inhibitor, had the broadest and most significant effect on cell viability and apoptosis in primary and recurrent/metastatic HNSCC-derived cells. Sensitivity to MK-1775 depended on several factors including the mutational status of TP53 , HPV status, and concomitant cisplatin use. TP53- mutant cell lines were more sensitive to MK-1775 than TP53 wild-type cell lines, and the use of MK-1775 combined with cisplatin enhanced the response when compared with cisplatin alone. Based on these preclinical results, several clinical trials are evaluating an MK-1775 monotherapy or a combination of MK-1775 and platinum chemotherapy ( NCT02196168 , NCT01748825 , and NCT02341456 ).

( A, B ) p53-dependent DNA damage response. When there is DNA damage, ATR and ATM initiate p53-dependent cell cycle arrest. In p53 mutant tumors, there is increased dependence on the G2-M cell cycle check point regulation; therefore, CHK1 inhibitors (AZD7762 and LY2606368) and a Wee inhibitor (MK-1775 or AZD-1775) are thought to be more effective in p53 mutant tumors. ( C ) Palbociclib is a selective CDK4 and CDK6 inhibitor and P276-00 inhibits CDK1, CDK4, and CDK9. ( D ) Olaparib and veliparib inhibits poly-ADP ribose polymerase (PARP) 1, which is a key driver of base excision repair on a single-strand break. CDK, cyclin-dependent kinase.

PI3K-mTOR targeted agents

The PI3K-mTOR pathway regulates numerous cellular functions including cell cycle, survival, metabolism, motility, genomic instability, angiogenesis, and inflammatory cell recruitment. The PI3K-mTOR pathway is the most frequently altered pathway in human tumors, with PIK3CA and PTEN being the most frequently altered genes in the pathway; therefore, enumerable therapeutic agents targeting this pathway are being developed. In HNSCC, approximately 30% of both HPV-positive and -negative tumors have genomic alterations in the pathway. Current data suggest that deregulation of this pathway plays an important role in development of HNSCC and response to PI3K-mTOR inhibitors.

Because of the prodigious preclinical evidence that PI3K-mTOR signaling represents an integral component of HNSCC signal transduction, several clinical trials are currently underway to evaluate the efficacy of small molecules that inhibit key points of this pathway ( Table 1 ). For example, PI3K inhibitors (BYL-719 and buparlisib) and mTOR inhibitors (everolimus, temsirolimus, sirolimus, and ridaforolimus) are actively being investigated in phase II trials as a single agent or in combination with previously established radiation and chemotherapy regimens in HNSCC. Because PI3K and mTOR share several structural similarities, some compounds have been developed to inhibit the class I PI3K isoforms and mTORC1/2, and several of these dual PI3K/mTOR inhibitors are in clinical development including NVP-BEZ235 and LY3023414. Metformin, a biguanide commonly used to manage type II diabetes, is also being investigated as a chemotherapeutic in HNSCC. Although it is not a direct mTOR inhibitor, metformin indirectly inhibits mTORC1 by increasing intracellular AMP levels mediated by AMPK-dependent and independent mechanisms.

Contrary to the perception that the presence of PIK3CA mutations would highly correlate with response to PI3K inhibitors, the trial evaluating cetuximab or cetuximab and PX-866 combination showed that the PIK3CA mutations were not associated with response in both arms. In addition, a recent study revealed that HNSCC cell lines with PIK3CA mutations or amplification were initially more sensitive to a PI3Kα isoform-specific inhibitor, BYL719, compared with the cell lines with wild-type PIK3CA . Acquired resistance, however, emerged through an interaction among the EGFR, Axl, and PI3K pathway and subsequent downstream activation of mTOR. The addition of mTOR inhibitor (RAD001 and AZD8055) resensitized the BYL719-resistant cells, confirming that the acquired resistance to the PI3K inhibition was mediated through maintained activity of mTOR. Furthermore, the combined inhibition of PI3Kα and either EGFR or Axl reversed the resistance, thereby showing that the maintained mTOR activity was through the interaction between EGFR and Axl and downstream activation of PKC. This study provided the scientific rationale to combine PI3K inhibitors with EGFR, Axl, or PKC inhibitors. The next step in the development of PI3K-mTOR inhibitors is to identify better predictive biomarkers of clinical benefits and to further understand the resistance mechanisms to improve the therapeutic index through appropriate combination regimens.

Notch pathway targeted agents

The NOTCH pathway comprises of four receptors (NOTCH 1–4) and two families of ligands: JAG1/2 (Jagged 1 and 2) and DLL1/3/4 (Delta-like 1, 3, and 4). The pathway is initiated when one cell expressing the appropriate ligand (Jagged or Delta-like) interacts with another cell expressing a NOTCH receptor. These cell surface receptors are expressed by various cell types and are generally involved in multiple biologic functions including differentiation, regulation of self-renewal capacity, cell-cycle exit, proliferation, angiogenesis, and survival. The importance of this pathway in HNSCC was demonstrated by whole-exome sequencing studies that reported 10% to 15% of HNSCC harbor NOTCH1 mutations. A prognostic significance of the mutation was reported by Song and colleagues that patients with NOTCH1 mutations had a significantly shorter overall and disease-free survival compared with those whose tumors carried no NOTCH1 mutation in a sequencing study of 51 oral squamous cell carcinomas. In addition, all 15 patients with lymph node metastasis and NOTCH1 mutations had tumor relapse or metastasis after attempted curative treatment, and 14 (93%) of the patients died during the follow-up period compared with only 2 (15%) of the 13 patients with no lymph node metastasis and no NOTCH1 mutation.

Another study examined the comprehensive genetic, epigenetic, and transcriptional alterations of NOTCH signaling in a cohort of HNSCC patients. Overexpression of downstream Notch effectors was seen in 32% of HNSCC (44 HNSCC tumors and 25 normal mucosal samples). When DNA copy number, methylation, and gene expression of the 47 NOTCH signaling pathway genes were evaluated, a bimodal pattern of NOTCH pathway alterations emerged. A small subset exhibited inactivating NOTCH1 mutations, but a larger subset exhibited other NOTCH1 pathway alterations including increased gene copy number of the receptors or ligands, increased gene expression, and downstream pathway activation. These findings suggested that the NOTCH pathway is a potential therapeutic target in a subset of HNSCC. As seen in these studies, NOTCH pathway is either tumor suppressive or oncogenic in HNSCC depending on the genetic context; therefore, development of biomarkers will enable the appropriate subset of patients to be treated with NOTCH inhibitors in clinical development.